Blood cardiolipin in haemodialysis patients. Its implication in the biological action of platelet-activating factor

Int. J. Biochem. Cell Biol. Vol. 28, No. 1, pp. 4351,

Pergamon

1357-2725(9!5)00114-6

1996 Copyright 0 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved 1357-2725/96 915.00 + 0.00

Blood Cardiolipin in Haemodialysis Patients. Its Implication in the Biological Action of Platelet-activating Factor SMARAGDI ANTONOPOULOU,’ CHRISTOS IATROU*

CONSTANTINOS A. DEtiOPOULOS,‘*

‘University of Athens, Department of Chemistry, Panepistimioupolis, GR 157-71 Athens and ‘General Hospital of Athens, Department of Nephrology, Messogion 154, GR 115-27 Athens, Greece Bovine heart cardiolipin spee&.ally inhibits rabbit platelet aggregation induced by PAF in uitro. In tbe past we have reported that patients with primary glomerukmephritis have hrcreasedPAF levels in plasma (Iatrou et a& 199Sb). In this work we huestigate the existence of cardiolipiu in tbe blood of end-stage renai pati~ts due to primary gkunemIonepbritis and we study its hnplicatioa in tife biological study of PAP. Lipids from biood sampiesof e&stage renal patients were extracted, fractiouated onto siiicic asid cohunu and onto IBgh Performauce Liquid Chromatography (HPLC) cation exchangecolumn. PAF fraction was removed aud -his were separated from the rest lipid fraction with current cormter distribution and Wtbermore fractionated onto I-IPLC silica column. Tbe results show: 1. cardioIipiu is present in the biood of end-stage renal patients. 2. Blood cardiolipin specitkaiiy iuhibii PAP-huiaced aggregation in washed rabbit platelets. 3. Scatchard plot analysisof PAP binding, iu the preseuceof mdabelied PAF and in the presenceof cardioiipin, shows that rabbit piatelets possesstwo diifereut types of biudhtg sites. One of which is saturable and of high affinity, kD = 0.103 + 0.03 uM (SEM, a =3) with 337 + 94 binding sites per platelet for PAF and kD=O.o87 + 0.021&l with 371 f 92.7 binding sites per piatelet for cardioiipin while the other one has ahnost h&rite bhulhrg capacity. 4. Blood cardioiipin competesi%BPAI?binding in rabbit platelets. This work showsthat cardioiipin exists in the biood of end-stage renai patients and speciiicaliy iuhiiits PAP-huiuced aggregation as well as PAF bhrding in rabbit platelets. The possibleimplication of the biiIogicai actions of cardioiipin in the anticardiolipin-autiphosphoIipid syndrome is aiso discussed. Keywords: Human Mood cardioiipia Platelet-activating factor lipid (anticardiogpiu) syndrome Haemodialysis

PA&receptors

Aatiphospho-

Int. J. Biochem. Cell Biol. (1996) 28, 43-51

range of effects both in vivo and in vitro. In addition, PAF can stimulate various cells, such as platelets, neutrophils, macrophages, endothelial cells and protozoan cells (Benveniste and Pretolani, 1985; Lee and Snyder, 1986; Tselepis et al., 1986). During the last ten years, various synthetic and naturally occurring compounds have been identified to act as specific PAF inhibitors (Saunders and Handley, 1987). Endogenous lipid inhibitors of PAF have been found in mammalian tissues and organs (Masugi et al., 1988; Miwa et al., 1987; Nakayama et al., 1987; Smal and Baldo, 1989). Bovine heart cardiolipin

INTRODUCI’ION

Platelet-activating factor (PAF), 1-0-alkyl-2acetyl-sn -glycero-3-phosphocholine (Demopou10s et al., 1979; Benveniste et al., 1979) is a naturally occurring lipid mediator with a wide *To whom all correspondence should be addressed. Abbreuiatiom: PAF, Platelet-activating factor (I-O-alkyl2-acetyl-sn-glycero-3-phosphocholine); HPLC, Highperformance liquid chromatography; BSA, Bovine serum albumin; ADP, Adenosine diphosphate; ESMS, Electrospray Mass Spectroscopy; SLE, Systemic lupus erythymatosus. Received 6 April 1995; accepted 22 August 1995. 43

44

Smaragdi Antonopoulou

is known to specifically inhibit rabbit platelet aggregation induced by PAF in vitro (Tsoukatos et al., 1993). Cardiolipin is a constituent (2-10% of the total phospholipids) of mammalian tissues and is characteristically associated with subcellular membranous particles, particularly with mitochondria, having an important biological role (Ioannou and Golding, 1979). In addition, circulating anticardiolipin antibodies are present in the blood of haemodialysed patients suggesting the existence of cardiolipin. Results from previous experiments (Antonopoulou et al., 1992), that compared blood samples examined by silicic acid column chromatography + HPLC with the ones examined without HPLC, indicate that omission of the HPLC purification step of blood extracts leads to lower PAF levels by almost two orders of magnitude than those obtained after further HPLC purification of the samples. These results suggest the existence of PAF inhibitor(s) in the fraction of silicic acid column chromatography which contains PAF. In this study an attempt is made to isolate and characterize specific PAF inhibitors and more importantly to investigate the existence of cardiolipin in the blood of end-stage renal patients due to primary glomerulonephritis as well as study its implication in the biological action of PAF. MATERIALS

AND METHODS

Biological Jruidr Blood samples (75 ml) were collected from end-stage renal patients who have been undergoing haemodialysis with primary kidney disease chronic glomerulonephritis. These samples were drawn from the efferent lines of the dialyser during the haemodialysis session, previously used for the determination of PAF levels in another research (Iatrou et al., 1995a). Materials and reagents All reagents were of analytical grade, purchased from Merck (Darmstadt, Germany). HPLC solvents were purchased from Ruthburn (Walkerburn, Peebleshire, U.K.). Lipid standards of HPLC grade were obtained from Supelco (Bellefone, Pa., U.S.A.). Semisynthetic PAF (80% C-16PAF and 20% C-18PAF) was synthesized in our laboratory as previously described (Demopoulos et al., 1979). [‘H]PAF in

et al.

ethanol solution was purchased from New England Nuclear (Boston, Mass.) with a spec. act. 10.0 Ci/mmol. Bovine serum albumin adenosine diphosphate (ADP), (BSA), thrombin and bovine heart cardiolipin sodium salt were purchased from Sigma (St Louis, MO., U.S.A.). Chromatographic material used for column chromatography was silicic acid 35-70 mesh ASTM 7733 (Merck, Darmstadt, Germany). Silicic acid column preparation Silicic acid was washed with water and methanol and activated overnight at 120°C. The glass column 1Omm (i.d.), was slurry packed (20 cm height) using methanol/water (1: 1.5, v/v). The lipid sample was applied in 1-2 ml of chloroform/methanol (1: 1, v/v). Instrumentation Column chromatography was performed in a glass column, 45 cm x 10 mm (i.d.). HPLC was performed on a dual pump Jasko (Tokyo, Japan) model 880-PU HPLC, supplied with a 330 ~1 loop Rheodyne (P/N 7125-047) injector. A Jasko model 875 U.V. spectrophotometer was used as detector (210nm). The spectrophotometer was connected to a Hewlett-Packard (Avondale, Pa, U.S.A.) model HP-3396A integrator-plotter. A cation exchange column was used, SS 10 pm Partisil, 25 cm x 4.6 mm (i.d.), PXS lo/25 SCX from Whatman (Clifton, NJ, U.S.A.) as well as an absorption column, Silica 25 cm x 4.6 mm (i.d.), from Hichrom H5. The flow rate was 1 ml/min. The PAF-induced aggregation was measured in a Chrono-Log (Havertown, Pa, U.S.A.) aggregometer coupled to an Omniscribe recorder (Houston, Tex., U.S.A.). Radioactivity was measured in a 1209 RackBeta Flexivial beta-counter (LKBPharmacia, Turku, Finland). Biological assays The platelet aggregation induced by PAF (1.25 x IO-” M, final concentration) which was inhibited by the examined sample was measured as PAF-induced aggregation, in washed rabbit platelets before and after the addition of the examined sample (Demopoulos et al., 1979). PAF and the examined samples were dissolved in 2.5 mg BSA per ml saline. This experiment was also performed with thrombin (0.125 U/cuvette), as well as with ADP, (1 PM, final concentration) in order to assess the

Implication of human blood cardiolipin

inhibition of thrombin or of the ADP-induced aggregation. Thrombin and ADP were dissolvedin saline. Binding assay

All binding assays were performed at 20°C using centrifugation as described elsewhere (Antonopoulou, 1994). The experimental conditions (number of washed rabbit platelets, buffer and PAF final concentration) for the incubation were the same as in the biological assay with the exception of temperature (20 instead of 37°C). The platelets were incubated for 5 min which was found to be the optimum time. In brief: platelet suspension 0.5 ml (1.25 x lob9 cells/ml) was incubated with increasing concentrations of [‘HI PAF (0.36 5.36nM) in the presence or absence of unlabelled PAF (2.5 x 10W8M) as well as in the presence of the examined samples (10 times the 50% inhibitory concentrations, I&,), both added 30 set before [3H] PAF addition. The bound rI-Il PAF was separated from the free [3w PAF by centrifugation in Ependorff tubes and the radioactivity was measured. The non-specific binding was determined in the presence of unlabelled PAF. The specific binding was calculated by subtracting the nonspecific binding from the total binding. For Scatchard plot analysis, the ratio of plateletbound (B) to free radioactivity (F) was calculated. Mass spectroscopy (MS)

The purified bioactive phospholipid was subjected to mass spectroscopy analysis. The Electrospray Mass Spectroscopy (ESMS) was recorded on a Fisons VG Quattro instrument with a VG Biotech Electrospray source, having a hexapole lens. Nitrogen 99.99% pure was used as the nebulizing and bath gas at flows of 20 and 150 dm’ min-’ respectively. The sample was injected in the flow of solvent (10 pl/min) of a Varian 9012 solvent delivery system, via a Fisons interface with a Reodyne 7125 injector. The capillary voltage was optimum at about 3 kV for negative ions and 3.30 kV for positive ions. The high voltage lens potential was kept at 0.56 kV. The focus and skimmer lens voltages were 40 and 45 V, respectively, in the majority of the measurements, as these values produced the highest peak intensities and minimum fragmentation. HPLC grade methanol/water (70: 30, v/v) 5% in ammonia was used as solvent.

in PAF actions

45

Statistical analysis

The Two-Sample Analysis procedure was used in order to estimate and test the means and variances of the two samples. Experimental procedure

Blood samples (2ml), immediately after collection, were poured into 20 ml of absolute ethanol in a 50 ml glass centrifuge tube. Consequently, these samples were subjected to the procedure of PAF isolation which has been described previously (Demopoulos et al., 1994). Briefly, after thorough mixing, the tubes were centrifuged at 1OOOg for 20 min. The supernatant was collected by decantation and after extraction with the Bligh-Dyer method (Bligh and Dyer, 1959), the “free” PAF extract was obtained. The pellet remaining in the 50 ml centrifuge tube from the blood samples was extracted as above and the “bound” PAF extract was obtained as well. Each of the above extracts were fractionated on a silicic acid column. The column was eluted with 45 ml of methanol/water, 1: 1.5 (v/v), followed by 50 ml of methanol/water 2 : 1 (v/v) and 40 ml of chloroform/methanol/water 1: 2 : 0.8 (v/v). The first 45-ml fraction and the last 40-ml fraction from silicic acid column chromatography were tested for PAF inhibitory effect. The first 45 ml (containing the bulk of proteinaceous and other non-lipid impurities) and the last 40 ml (containing most of the lipids) were discarded. The PAF containing eluents (50 ml) were phased by adding chloroform and methanol to a final ratio of chloroform/methanol/water of 1: 1: 0.9 (v/v/v). The PAF containing lower layer was evaporated until dryness. The dry residue of the previous step was further fractionated onto a cation exchange HPLC column using acetonitrile 60% and methanol/water (4: 1) 40% (v/v) as eluting solvents (Andrikopoulos et al., 1986). The fraction where PAF was eluted (approx. 22-24 min), was collected and used in another research (Iatrou et al., 1995a) while the eluents before PAF fraction from both extracts (“free” and “bound”) were pooled together. The lipid fraction was separated in neutral lipids and phospholipids by current counter distribution (Galanos and Kapoulas, 1962). The phospholipids were further fractionated onto a HPLC silica column eluted with acetonitrile from 0 to 10 min, followed by a linear

46

Smaragdi Antonopoulou

gradient to methanol in 10 min, and again followed by a second hold in methanol in 15 min. All the eluted lipid fractions were tested for their ability to inhibit PAF, thrombin and ADPinduced aggregation in washed rabbit platelets. The fractions which inhibited PAF-induced aggregation were subjected to binding assay. RESULTS

As it was expected, we did not detect any inhibitory activity in the first (45 ml methanol/water, 1: 1.5 v/v) nor in the last fraction (40 ml chloroform/methanol/water, 1: 2: 0.8 v/v) of silicic acid column chromatography, which were then discarded. On the other hand, we detected strong inhibitory activity against PAF-induced aggregation in the eluents before PAF fraction of the cation exchange HPLC column. It must be mentioned that in order to result to 50% inhibition of the PAF-induced aggregation (IC,,), an amount corresponding to 0.25 ml of the initial blood sample (75 ml) was needed. These eluents were pooled together and were subjected to current counter distribution. The fraction of phospholipids, as well as the fraction of neutral lipids, were tested for their ability to inhibit PAFinduced aggregation in washed rabbit platelets. The PAF-inhibitory activity expressed as IC,, was for phospholipids and neutral lipids an

et al.

amount corresponding to 0.083 ml and 0.5 ml respectively of the initial blood sample. These data indicate that the majority of the inhibitory effect was due to phospholipids while a minor amount was due to neutral lipids. The fraction of phospholipids was further fractionated onto a HPLC silica column and all the eluted lipid fractions were tested for their ability to inhibit PAF-induced aggregation in washed rabbit platelets. From the above experiment, we detected a lipid which was eluted in the area of standard cardiolipin (Fig. 1) and which strongly inhibited PAF-induced aggregation; an amount corresponding to 0.1 ml of initial blood sample was needed in order to obtain 50% inhibition of PAF-induced aggregation. On the other hand, this lipid did not effect thrombin and ADP-induced aggregation in washed rabbit platelets. Consequently, [3H] PAF binding assay was studied in both absence or presence of unlabelled PAF as well as in the presence of the lipid with cardiolipins’ retention time. Under the above conditions, PAF binding was increased concentration-dependently of the amount of added labelled PAF (Fig. 2). Essentially, the same increase was observed in the presence of unlabelled PAF as well as in the presence of the active lipid. Specific binding defined as total E3H]PAF binding minus unspecific binding in the presence of unlabelled ligands

206 nm 0.4 a.u.f.s. MeOH

100%

active lipid

15

25

time (min) Fig. 1. Fractionation of phospholipida onto a PPLC silica column. The retention time of standard cardiolipin was from 22.5 to 24.5 min, approx. The retention time of the active lipid was at 23.5 min. Elution system as indicated. ACN: acetonitrile; MeOH:methanol.

Implication of human blood cardiolipin

in PAF actions [3H] PAF

[3 H] PAF + clrdiolipin

5

BH]PAF (nM)

Fig. 2. rH] PAF binding to intact rabbit platelets in the absence (0) or presence (0) of unlab4ed PAF (2.5 x IO-* M) or in the presence (A) of cardiolipin (ten times the 50% inhibitory concentration). The non-specific binding was determined in the presence of unlabelled PAF (0) or in the presence of cardiolipin (A). Specific binding of PAF (m) and specific binding of cardiotipin (A) were calculated by subtracting the corresponding non-specific binding from the total binding. Values are means of three experiments.

reached a plateau at 0.7 nM rIIj PAF. The calculated specific PAF binding to 1.25 x 10’ platelets was 70 & 18.9 fmol in the presence of unlabelled PAF and 77.5 If: 19 fmol in the presence of the active lipid (values are means of three experiments, P = 0.65). Scatchard plot analysis (Fig. 3) of PAF binding in the presence of unlabelled PAF and in the presence of the active lipid showed that rabbit platelets possess two different types of binding sites. One of these was saturable and of high affinity (kD = 0.103 + 0.03 nM with 337 f 94 binding sites per platelet for PAF and kD = 0.087 + 0.02 nM with 371 + 92.7 binding sites per platelet for the lipid, P = 0.48 and P = 0.68, respectively) and other the

one had nearly infinite binding capacity (kD = 0.87 f 0.24 nM with 1150 f 322 binding sites platelet for PAF and kD = per 0.87 + 0.22 nM with 1150 f 287.7 binding sites per platelet for the lipid, P = 1). The statistical analysis showed that the specific binding of PAF on rabbit platelets did not differ from the specific binding of the active lipid on the same platelets, whereas the non-specific binding for both compounds (PAF and the active lipid) were essentially the same. These results are in accordance with the ones already reported for PAF binding in rabbit platelets (Inarrea et al., 1984; Homma et al., 1987). The identification of the active lipid with ESMS analysis (Figs 4-6) showed that this lipid was an analog of cardiolipin (MW: 1448), with four similar fatty acids (linoleic acids, 18:2). Positive ion spectra showed peaks at m/z 1466, corresponding to [M + 18]+ due to the presence of ammonia in the solvent system. Negative ion spectra show peaks at m/z 1447, corresponding to [Ml-, 723, corresponding to w12- and 279, corresponding to the carboxylate ion. DISCUSSION

The existence of endogenous PAF inhibitors in various mammalian cells and tissues has been

of platelet-bound (B) to free radioactivity (F) was calculated. Values are means of three experiments.

were provided Phospholipids,

for their chemical structures. with a chromatographic

48

Smaragdi Antonopoulou

er al.

101

14c7.2

Fig. 4. Positive ion ESMS spectrum of the active lipid.

behaviour similar to that of cardiohpins and phosphatidylglycerols, have been isolated from guinea pig heart and specifically were found to inhibit PAF-induced aggregation in vitro (Tsoukatos et al., unpublished observations). It is also known that bovine heart cardiohpin 101

inhibits PAF-induced aggregation in washed rabbit platelets but has no effect on ADP or arachidonic acid-induced aggtegation (Tsoukatos et al., 1993). To our knowledge there is no reference in the literature describing the existence of cardiofipin 723.4

,723.g

7244

9

Fig. 5. Negative ion ESMS spectrum of the active lipid, m/z ranging from 1013to 800.

Implication

of human blood cardiolipin

49

in PAF actions

24.1

'24.6

1448.6 125.7

we

Fig. 6. Negative ion ESMS spectrum of the active lipid, m/z ranging from 700 to 1600.

in human blood. In the present study, we investigated the possible existence of a specific PAF inhibitor in end-stage haemodialysed glomerulonephritis patients. We isolated a phospholipid which has been identified as an analogue of cardiolipin since its retention time in HPLC is identical to the one of standard cardiolipin and also the MS analysis shows characteristic fragments of cardiolipin. This phospholipid specifically inhibited PAFinduced aggregation in washed rabbit platelets with a dose-dependent manner. The present results correspond well to those described in the literature (Tsoukatos et al., 1993) of cardiolipin inhibition of PAF-induced rabbit platelet aggregation. In this study we also investigated the competition of the isolated phospholipid with PAF binding sites in washed rabbit platelets. A direct competition of this phospholipid with PAF binding sites can be proposed since this phospholipid displaced platelet-bound t31-IlPAF in a concentration-dependent manner as compared to unlabelled PAF. In addition, antiphospholipid antibodies also have been detected in high levels in the serum of patients with end-stage renal failure who have been undergoing haemodialysis (from these patients, the highest -serum levels were foundin those having glomerulonephritis as a primary disease) (Kirschbaum et al., 1991; Sitter et al.,

1992) and in these patients, a positive correlation between the serum anticardiolipin antibodies and the frequency of the vascular thrombosis (arterio-venous Ilstula or elsewhere) was detected. The precise explanation for the existence of the circulating high levels of antiphospholipid antibodies in haemodialysis patients and their pathophysiologic effect has not been fully elucidated. Although several theories exist regarding the involvement of the antiphospholipid antibodies in the production of vascular thrombosis, the exact way in which these antibodies act is not exactly known. According to the most predominant theory, there is a cross-reaction between the antiphospholipid antibodies and the phospholipids in the endothelial cells membrane which leads to the reduction of arachidonic acid release and prostacyclin production (Tsakiris et al., 1989). Simultaneously, these antibodies prevent the action of antithrombin III and proteins C and S, while their binding with the negative charged phospholipids of the platelet membrane activates platelets, promoting their aggregation and thrombus production (Lo et al., 1990; Ruiz-Argue&s et al., 1991). However, participation in vascular thrombosis in patients with antiphospholipid antibodies has also been observed as a result of impaired fibrinolysis (Tsakiris et al., 1989), since lupus

Smaragdi Ann mopoulou et al.

50

anticoagulant demonstrated that ,&hey have a selective inhibitory action on thrombomodulin function which normally induces the activation of protein C and prevents the normal fibrolytic mechanism. Since it was demonstrated that (a) cardiolipin is a PAF specific inhibitor, (b) PAF induces platelet aggregation as well as coagulation system activation (Silvestro et al., 1994; Ou et al., 1994), it is possible that the observed vascular thrombosis in haemodialysed patients as well as in the anticardiolipin or antiphospholipid syndrome is a result of the inactivation. of cardiolipin from the circulating antieardiolipin antibodies which leads to higher PAF activity and it is not a result of the previously mentioned mechanism. High levels of PAF in the blood have been indeed demonstrated in patients with SLE (Love and Santoro, 1990; Harris el al., 1983) as well as in haemodialysis patients (Garcia-Martin et al., 1991; Kirschbaum et al., 1991; Sitter et al., 1992) and therefore the observed thrombosis in that group of patients could be the result of the PAF action on platelets and coagulation cascade. Acknowledgement-We greatly appreciate the assistance of Mr Dimitris Argyropoulos, Laboratory of Inorganic Chemistry, University of Athens, in performing the Electrospray Mass Spectroscopy.

REFERENCES Andrikopoulos N. K., Demopoulos C. A. and SiafakaKapadai A. (1986) High-performance liquid chromatographic analysis of platelet-activating factor on a cation-exchange column by direct ultraviolet detection. J. Chromatogr. 363, 412417. Antonopoulou S., Andrikopoulos N., Demopoulos C. A. and Kapoulas V. M. (1992) Simple and precise method for the determination of PAF in blood (B) and urine (U). In Fourth International Congress on PAF and Related Lipid Mediators, p. 41. Snowbird, Utah, U.S.A. Antonopoulou S. (1994) Study of compounds with PAF-like activity and PAF-inhibitors from natural (human blood and Urtica dioica) and syn&e.tic sources. Doctoral Dissertation, University of Athens. Athens, Greece. Benveniste J. and Pretolani M. (1985) PAF-acether (platelet-activating factor): its role in inflammation. Ado. Injlam. Res. 10, 7-19. Benveniste J., Tence M., Varenne P., Bidault J., Boullet C. and Polonsky J. (1979) Semi-synthese et structure proposee du facteur activant les plaquettes (PAF: PAFacether, un alkyl ether analogue de la lysophosphatidylcholine). C. R. Hebd. Seanc. Acad. Sci. Paris 289, 1037-1044X Bligh E. G. and Dyer W. J. (1959) A rapid method of total lipid extraction and purification. Can. J. Bioehem. Physiol. 37, 911-917.

Demopoulos C. A., Pinckard R. N. and Hanahan D. J. (1979) Platelet-Activating Factor. Evidence for I-O-alkyl2-acetyl-sn-glyceryl-3-phosphorylcholine as the active component (a new class of lipid chemical mediators). J. Biol. Chem. 254, 9355-9358. Demopoulos C. A. Andrikopoulos N. K. and Antonopoulou S. (1994) A simple and precise method for the routine determination of Platelet-Activating Factor in blood and urine. Lipids 29, 305-309. Galanos D. S. and Kapoulas V. M. (1962) Isolation of polar lipids from triglyceride mixtures. J. Lipid Res. 3, 134137. Garcia-Martin F., De Arriba G. and Carrascosa T. (1991) Anticardiolipin antibodies and lupus anticoagulant in end-stage renal disease. NephroZ. Dial. Transplant. 6, 543-547. Harris E. N., Gharavi A. E., Boey M. L., Pate1 B. M., Mackworth-Young C. G. and Loizou S. (1983) Anticardiolipin antibodies: detection by radioimmunoassay and association with thrombosis in systemic lupus erythematosus. Lancer 2, 1211-1214. Homma H., Tokumura A. and Hanahan D. J. (1987) Binding and internalization of platelet-activating factor l-0-alkyl-2-acetyl-m-glycero-3-phoaphocholine . washed rabbit platelets. d. Biol. Chem. %?, 10582-1058~ Iatrou C., Afentakis N., Antonopoulou S., Andrikopoulos N., Demopoulos C. A. and Ziroyiannis P. (1995a) The production of Platelet-Activating Factor (PAF) during hemodialysis with cuprophane membrane. Does calcium concentration in the dialysate play any role on it? Znt. J. Artif. Organs, 18, 355-361. Iatrou C., Moustakas G., Antonopoulou S., Demopoulos C. A. and Ziroyiannis P. (199Jb) Platelet-Activating Factor (PAF) levels and PAF Acetylhydrolase (AH) activities in patients with primary gIomerulonephritis (PGN). Nephron, in press. Inarrea P., Gomez-Cambronero J., Nieto M. and Crespo M. S. (1984) Characteristics of the binding of platelet-activating factor to platelets of different animal species. Eur. J. Pharmacol. 105, 309-315. Ioannou P. V. and Golding B. T. (1979) Cardiolipins: their chemistry and biochemistry. Prog L.&id Res. 17,279-3 18. Kirschbaum B., Mullinax F., Curry N. and Mallory J. (1991) Association between anticardiolipin antibody and frequent clotting problems in hemodialysis patients. J. Am. Sot. Nephrol. 2, 332. Lee T.-C. and Snyder F. (1986) Function, metabolism and regulation of Platelet-Activating Factor and related ether lipids. In Phospholipti and Cellular Regufation (E&ted by Kuo J. F. and Boca R.), Vol. 2, pp. l-39. CRC Press Inc., Boca Raton, Fla. Lo S. C., Salem H. H. and Howard M. A. (1990) Studies of natural anticoagulant proteins and anticardiolipin antibodies in patients with the lupus anticoagulant. Br. J. Haematol. 76, 380-386. Love P. E. and Santoro S. A. (1990) Antiphospholipid antibodies: anticardiolipin and the lupus anticoagulant and systemic lupus erythematosus (SLE) and in non disorders. Prevalence and clinical significance. Ann. Intern. Med. 112, 682-698. Masugi F., Orihara T., Saeki S., Sakagushi K., Kumahara Y., Satouchi K., Oda M., Saito K. and Tokunaga K. (1988) Endogenous platelet-activating factor and antiplatelet-activating factor in patients with renovascular hypertension. Life Sci. 42, 455-460.

Implication of human blood cardiolipin in PAF actions Miwa M., Hill C., Kumar R., Sugatani J., Olson M. S. and Hanahan D. J. (1987) Occurrence of an endogenous inhibitor of platelet-activating factor in rat liver. J. Biol. Chem.

262, 527-530.

Nakayama R., Yasuda K. and Saito K. (1987) Existence of endogcnous inhibitors of Platelet-activating Factor (PAF) with PAF in the rat uterus. J. Biol. Chem. 262, 1317413179. Ou M. C., Kamboyashi J., Uemura Y., Shinazaki K., Shibou E., Sakon M., Yukawa M. and Mori T. (1994) Potential etiologic role of PAF in two major septic complications: disseminated intravascular coagulation and multiple organ failure. Throm. Res. 73, 227-238. Ruiz-Arguelles G. J., Ruiz-Arguelles A. and AlarconSegovia D. (1991) Natural anticoagulants in systemic lupus erythematosus. Deficiency of protein-S bound to C4bp associates with recent history of venous thromhoses, antiphospholipid antibodies, and the antiphospholipid syndrome. J. Rheumatol. 18, 552-556. Saunders R. N. and Handley D. A. (1987) Platelet-Activating Factor antagonists. Ann. Rev. Pharmacol. Toxicol. 27, 237-255. Silvestro L., Ruikun C., Sommer F., Due T. M., Biancone L., Montrucchio G. and Camussi G. (1994) Plateletactivating Factor induces endothelial cell expression of

51

adhesion molecules and modulation of surface glycocalyx evaluating by electron spectroscopy chemical analysis. Sem. Thr. Haem. Zo, 214-227. Sitter T., Sparmagl M. and Schil’R H. (1992) Anticardiolipin antibodies and lupus an&a gulant !n pa&& &a&d with different methods of renal replacement therapy in comparison to patients with systemic lupus erythematoSW

Ann.

Hematol.

65, 79-82.

Smal M. A. and Baldo B. A. (1989) Inhibitors of PAF in human saliva. In Third International Conference on Platelet Activating Ether Lipids, p.

Factor

and Structurally

Related

Alkyl

132. Tokyo, Japan. Tsakiris D. A., Marbet G. A., Makris P. E., Settas L. and Duckert F. (1989) Impaired fibrinolysis as an essential contribution to thrombosis in patients with lupus anticoagulant. Thromb. Haemost. 61, 175-177. Tselepis A., Tsoukatos D., Demopoulos C. A. and Kapoulas V. M. (1986) Effects of AGEPC on the intracellular levels of ions in Tetrahymena pyriformis. Biochem. Int. 13, 999-1008. Tsoukatos D., Demopoulos C. A., Tselepis A. D., Moschidis M. C., Donos A., Evangelou A. and Benveniste J. (1993) Inhibition by cardiolipins of PlateletActivating Factor-induced rabbit platelet activation. Lipids 28, 1119-I 124.